In [15]:
%%time
print(datetime.datetime.now())
print('About the low score in the southern ocean and SSP585\n'+\
'-> Maps of score and 3 grid-points chosen: fig')
# Load data2plot
savedfile = dirout + 'data2plot-score-maps-and-timeseries.pckl'
with open(savedfile, 'rb') as f1: data2plot = pickle.load(f1)
print("File loaded: "+savedfile)
#-----------------
# FIGURE PARAM
#-----------------
cm2in = 1/2.54 # converting factor cm to inch
nrow = 1
ncol = len(simu_list)
ccrsproj = ccrs.Robinson(central_longitude=-155)
fsize = (18*cm2in*2, 16*cm2in*2) # width height
fig, ax = plt.subplots(nrow, ncol, figsize=fsize,
subplot_kw=dict(projection=ccrsproj),
squeeze = False)
subnum = list('abcdefghijklmnopqrstuvwxyz')
aaa=[]
for sss in subnum: aaa.append('a'+sss)
subnum=subnum+aaa
subnum.reverse()
#-----------------
# CREATE CUSTOM CMAP
#-----------------
cmap = cm.get_cmap('viridis', 256)
newcolors = cmap(np.linspace(0, 1, 10))
cmap = ListedColormap(newcolors[:])
cmap.set_under(color='silver')
cmap.set_over(color='silver')
cmap.set_bad(color='silver', alpha=0)
#-----------------
# KEYWORDS DICT
#-----------------
kwmap = {'vmin':0, 'vmax':1, 'cmap':cmap, \
'transform':ccrs.PlateCarree() }
kwscatter = dict(edgecolor='k', facecolor='lightgray', alpha = .8, s=60, \
transform=ccrs.PlateCarree())
kwtrue = dict(lw=1, ls='-' , label='target' , color='k' )
kwpred = dict(lw=1, ls='--', label='predicted' , color='darkorange')
kwdiff = dict(lw=1, ls='-' , label='target - pred.', color='limegreen')
kwtrnd = dict(lw=1, ls='-' , label='trend + seas.' , color='silver')
kwbbox = dict(boxstyle ="round", fc ="0.8")
#=====================
#=====================
# MAPS
#=====================
#=====================
irow = 0
icol = 0
for simu in simu_list:
zax = ax[irow, icol]
X = data2plot[simu]['score']['X']
Y = data2plot[simu]['score']['Y']
Z = data2plot[simu]['score']['Z']
pcm = zax.pcolormesh(X, Y, Z, **kwmap)
zax.set_title(subnum.pop() + ') ' + simu.upper())
zax.coastlines()
# gl = zax.gridlines(crs=ccrs.PlateCarree(), linewidth=1)
for gridpt in gridpt_list:
jj, ii = gridpt[0], gridpt[1]
zax.scatter(X[jj, ii], Y[jj, ii], **kwscatter)
#
#
icol+=1
#
#=====================
#=====================
# TIMESERIES
#=====================
#=====================
axts = []
for isimu, vsimu in enumerate(simu_list):
for igridpt, vgridpt in enumerate(gridpt_list):
#---------------
# create axes
#---------------
zw1 = ax.flat[isimu].get_position()
nx0 = zw1.x0+.1*zw1.width
ny0 = zw1.y0 - (igridpt+1)*1.2*zw1.height
nw = .8*zw1.width
nh = .9*zw1.height
zaxts = fig.add_axes([nx0, ny0, nw, nh])
axts.append(zaxts)
#---------------
# Plot
#---------------
X = data2plot[vsimu]['timeseries '+str(igridpt)]['X']
Y = data2plot[vsimu]['timeseries '+str(igridpt)]['trdseas']
lltrnd, = zaxts.plot(X, Y, **kwtrnd)
Y = data2plot[vsimu]['timeseries '+str(igridpt)]['true']
lltrue, = zaxts.plot(X, Y, **kwtrue)
Y = data2plot[vsimu]['timeseries '+str(igridpt)]['pred']
llpred, = zaxts.plot(X, Y, **kwpred)
Y = data2plot[vsimu]['timeseries '+str(igridpt)]['diff']
lldiff, = zaxts.plot(X, Y, **kwdiff)
zaxts.axhline(0, lw=.5, c='k')
zwtxt = subnum.pop() + ') %.0fE, %.0fN' %data2plot[vsimu]['timeseries '+str(igridpt)]['coords'] + \
', score: %.2f' %data2plot[vsimu]['timeseries '+str(igridpt)]['score']
zaxts.set_title(zwtxt)
if isimu == 0: zaxts.set_ylabel('[mmol$\,$C$\,$m$^{-2}\,$d$^{-1}$]')
#
#
axts = np.array(axts).reshape((len(simu_list), len(gridpt_list))).T
#=====================
#=====================
# Decorate figure
#=====================
#=====================
#---------------------
# Colorbar
#---------------------
zw1 = ax[0, 0].get_position()
zw2 = ax[0, -1].get_position()
nx0 = zw1.x1
ny0 = zw1.y1 + .3*zw1.height
nw = zw2.x0 - nx0
nh = 0.04*nw
cax = fig.add_axes([nx0, ny0, nw, nh])
cbar = fig.colorbar(pcm, cax=cax, orientation='horizontal', \
ticklocation='top', extend='min')
cbar.set_label('Score of the kernel ridge regression for '+model, size=12)
#---------------------
# Set time series labels
#---------------------
for zax in axts[0:-1, :].flat:
zax.set_xticklabels('')
#
for zax in axts[-1, :].flat:
zax.set_xlabel('Years')
#
#-----------------
# Legend
#-----------------
zw1 = axts[-1, 0].get_position()
zw2 = axts[-1, -1].get_position()
nx0 = (zw1.x0 + zw2.x1)/2.
ny0 = zw1.y0 - .6*zw1.height
nw = 0
nh = 0
legax = fig.add_axes([nx0, ny0, nw, nh])
hdls = [xxx for xxx in axts[0, 0].lines]
hdls = [lltrue, llpred, lldiff, lltrnd]
legax.legend(handles=hdls, ncol=len(hdls), loc='center')
legax.axis('off')
#---------------------
# Save figure
#---------------------
fignam = 'score-maps-and-timeseries.png'
fig.savefig(dirout+fignam, bbox_inches='tight')
print('Figure saved: '+fignam)
# 23s
2023-06-28 15:10:36.360709 About the low score in the southern ocean and SSP585 -> Maps of score and 3 grid-points chosen: fig File loaded: dimred-230621-figure-article-review1-v0/data2plot-score-maps-and-timeseries.pckl Figure saved: score-maps-and-timeseries.png CPU times: user 21.9 s, sys: 1.57 s, total: 23.4 s Wall time: 21.8 s
